Comment by pnathan
9 years ago
This is very, very wasteful compared to actual mass transit. A subway network is much more effective at delivering people.
If he's looking for mega-good, Musk would do significantly better to drop a full subway network.
edit: tunnelling is a broadly solved problem. It's difficult, expensive, slow, etc. But there's no engineering reason why a hole in the ground can't happen. Musk might be able to drive some significant improvements there. No idea. But tunneling itself is not a reason to knock an idea beyond the cost and geoengineering involved.
You're asking the wrong question. There are over a hundred million Americans who commute long distances in their car. Mostly because they live/work in areas that aren't dense enough to have good last-mile public transit, and they don't want to deal with transfers. You can lecture them all you want, but ultimately, these people aren't going to downgrade their lifestyle just to fit your ideas of engineering efficiency.
The real question is: is this new solution more efficient than the current alternative of driving on the highway?
If places aren't dense enough to support mass transit, they are certainly not dense enough to financially support car-trains tunnels. The depicted tunnels have the same costs as building subways, and I imagine similar maintenance costs. On the other hand, they can only serve a fraction of the riders a traditional subway can, which means the cost per rider is huge. Not to mention it still requires owning a car, so the traditional capital and insurance savings from mass transit don't apply.
But to answer your question: maybe, but it doesn't matter. American cities are already bankrupting themselves in road maintenance. Adding a series of super car tunnels for an efficiency benefit? Out of the question.
Why does everyone lack imagination on this topic?
Problem: if you drive to the subway, then take the subway, then you have two problems: where do you park your car, and how do you get to your destination once you get off the subway if it's not walking distance?
Currently - people just drive. And the surface streets get progressively more crowded. We could drill more regular roadways - but it's notable that one reason we don't is controlling emissions and safety is difficult with human drivers.
So, taking that back to the video: replace the subway with general purpose transport stations that provide a mix of subway-like transport of passengers, or entire vehicles, powered electrically.
No vehicle emissions (huge problem with tunnels) and no endpoint transport issues - the subway becomes an extension of the road network, and if it's cheap enough to do, hopefully a very scalable one.
5 replies →
This solves the last Mile problem
2 replies →
> Mostly because they live/work in areas that aren't dense enough to have good last-mile public transit, and they don't want to deal with transfers.
Well, suburbanization was a totally dumbass move. Maybe instead of building tunnels, we could be rebuilding cities. We could be designing them to be attractive enough that people would want to live there.
You'd have to get the government to stop building roads...
which of course is seems crazy to most people. But the consequences are all around us.
1 reply →
When a city in America is destroyed by a nuclear explosion, Americans will re-learn why suburbia was so popular. Hopefully this won't happen for a long time and we will have a good stretch of city living. I love cities, but having millions of people concentrated enough to be killed by a single device; this is very different world than the one humans evolved in.
5 replies →
> : is this new solution more efficient than the current alternative of driving on the highway?
No.
Because - the exurbs and beyond simply don't have the density numbers to make this kind of investment pencil out without something like a 1000x class drop in costs, which would be wildly optimistic for physical equipment cost savings. Some suburbs might be able to handle it; Bellevue in the Puget Sound comes to mind immediately, but it's only a suburb in the context of Seattle; it'd be a major city in its own right in most of the US.
Further, you're not even getting to the fun part of driving a car - the wind, the sights, the open road. You've got a dang tunnel there. I'd get mildly claustrophobic and probably nauseous: subways already do that to me a little bit.
It's probably much more effective public policy at the federal level to focus on densifying American cities and reversing sprawl: this generates a nice sequence of network effects related to funding and infrastructural improvements. Among those would, eventually, be the demand for nice buses and nice trains with a regular security presence.
>Further, you're not even getting to the fun part of driving a car - the wind, the sights, the open road. You've got a dang tunnel there. I'd get mildly claustrophobic and probably nauseous: subways already do that to me a little bit.
I sure do love the wind, the sights, the open road of stop and go traffic every day.
Avoiding driving for ours in congestion would sound like an upgrade to me...
if you'd create 124mph transportation from 20-60 miles outside and through the city into the city, you wouldn't need to drive into the city. Most commuter trains average less than 50 so it's worth trying to drive. The average speed of the NYC subway is less than 35 mph. Light Rail?
Low-speed maglev: https://www.youtube.com/watch?v=c8SqDVUdMtY
What about waiting time between transfers? You need a certain population density to have almost continuous trains. In Bay Area, you have to wait an hour between Caltrains unless it is rush hour. Caltrains could go 500mph, I would still drive to avoid the wait between trains.
1 reply →
The video shows some sort of shuttle using the track. Those could be small busses or even public transport.
http://imgur.com/a/Wv8YD
In effect, this is a subway system. It's an underground rail system. The major difference is that you have a mixture of public and private passenger vehicles and smaller vehicles.
It's still incredibly wasteful compared to a real subway. The amount of people a real mass transit system can move is orders of magnitude above tons if individual little cars:
http://penguindreams.org/blog/self-driving-cars-will-not-sol...
..not to mention all the energy required to move each of those vehicles compared to a train that can move hundreds of more people in a similar space.
Americans need to get over all their train/bus hate. Other countries love mass transit. Many Americans try to shoot down any attempt to even put in a small system (small systems can grown) and kill off attempts to grow existing systems (see the Seattle Green Line).
Self driving cars can work great in Europe, where there is tons of transport and you just need to solve the last leg (or where self-driving trucks are rented just to move large items). In America we have massive gridlock due to a lack of rails and that needs to be fixed before self driving tech will fix anything.
>It's still incredibly wasteful compared to a real subway.
Because real subways have miserable sardine-can standing room conditions during commute times. Public transit is dead in the water in the US because its advocates use terms like "incredibly wasteful" to describe making systems anywhere near as comfortable as private cars.
For most people, the most comfortable chair they own is their driver seat. For many, the commute is the only time they get to be alone, meditative, and in complete control of their environment (cube farm or open office at work, children at home, etc).
When the CIA forces people to spend hours with their arms fully extended over their heads while bombarding them with 100db noise, there's a Senate inquiry. When BART does it, it's a regular Tuesday at 9am. (Yes, this is an extreme comparison, real stress positions are much worse, but your average Midwestern suburbanite used to his Toyota Camry is in for a real shock).
If you want Americans to get over their train/bus hate, then don't advocate such drastic reductions in the quality of our lives or the livability of our cities. Your average SUV-driving Wisconsin soccer mom has been to Manhattan as a tourist and decided that her one accidental peak-hours train ride was enough for one lifetime.
That, or densify the environment. We might put up with transit if the rides were shorter.
(I am a daily BART rider, and public transit dependence is the #1 reason I want to GTFO of the Bay Area).
7 replies →
One weakness of all public transportation is the lack of personal space. Maybe if someone would have solved that either by more space or some deep psychological insights, it would become more popular among Americans?
15 replies →
Doing some googling I found this as one of the links:
http://www.lamag.com/driver/l-s-subway-cars-will-definitely-...
It's a shame. Public transportation would be so awesome if it weren't for the bad press.
Is it really wasteful? It seems like it is an ever moving system. In terms of people capacity it seems it would carry more by being always flowing vs a stopping starting transit system.
9 replies →
According to the DOT, a train is only about 33% more energy efficient then driving.[1]
In addition a subway or bus system has to run on a schedule even when the train is mostly empty. The train and bus is always starting and stopping and causing all it's passengers increased delay to serve the needs of the few entering or leaving.
Public transport is the token ring networking of transportation. It's just inherently less efficient compared to a P2P routing system that only operates when there is a need and with the minimum number of stops (which is exactly 1).
[1]: https://www.rita.dot.gov/bts/sites/rita.dot.gov.bts/files/pu...
3 replies →
It's not a wasteful when US has huge land and those fuel vehicles need to travel a long distance may encounter car accidents e.g. fatigue, winter, floods, would be benefits from reduce air pollution in the these concept tunnel, in contrast, the size of Singapore is a lot smaller.
The tunnels could use solar energy which is not wasteful.
1 reply →
But... a subway for cars? This strikes me as significantly less efficient use of resources than everyone switching over to printing out their emails/IMs/photothings and faxing them to each other.
Edit: I mean, sure, great science fiction fun to think about!
yeah, the disparity in cross-sectional air displacement per capita is quite substantial. It is kind of funny to think about as an inefficient train. However, it does not need to replace trains or train commuters in any way. It only needs to contend with normal car travel.
The drag:person inefficiency is a price we still consider worth paying presently, even when it also requires an inefficient decentralized gas-burning energy conversion.
Add in reduced vehicle wear, and of course the time advantage, and with some of the efficiency gains like centralized power/reduced friction/huge electromagnetic actuators/maybe partially evacuated tunnel, it could still conceivably be profitable at a bargain on gas travel.
However, i think the advantages permit it to be sold for more than we are paying for gas travel. People already pay daily just to use a special lane which is nearly the same hassle with only slightly reduced traffic.
But can you get a moving block signalling system such that they can follow each other with minimal separation? Until you can do that, a real train will always win at people/hour due to the fact that you can have more people per consist as ultimately the signalling challenges are comparable and you can therefore have comparable separation.
With all my sympathy for metro - in many American cities it will never be good enough due to extensive suburbs.
In places where land is cheap, and there are single family houses, cars will be the most efficient way to commute for years to come. We'll switch to self-driving electric cars, but still.
5 replies →
64 LA subway cars cost $176 million [1]. That seems really high to me, at about $3 million per subway car.
Also people per hour is not really the metric the average person cares about. It's travel time to destination.
[1] http://www.lamag.com/driver/l-s-subway-cars-will-definitely-...
5 replies →
This is still completely ridiculous and wasteful.
> tunnelling is a broadly solved problem
As rockets were before Space X?
It costs a billion dollars per mile to dig a subway tunnel. If the costs could be brought down, it would change a lot about the way we build mass transit systems.
What? Are you serious??? If it's true then in US you are getting ripped off big time. The whole crossrail project is 73 miles with 14 miles of tunnels and 40 station and it has a cost of 15B£ (19B$ with today exchange rate). I cannot find the total cost of the tunnels, but apparently for 1.25B£ you can have 18km of tunnels: http://www.crossrail.co.uk/news/articles/crossrail-awards-ma... That is 1.6B$ for 11 miles. Basically an order of magnitude less than your figure of 1B$ per mile. If really Musk wants to improve by an order of magnitude over that figure then he is late, the crossrail tunnel have been completed 2 years ago.
Those Crossrail numbers give $0.2B/mile, including a lot of miles of track that aren't tunnels. After digging a bit I found this, which is a good summary:
https://pedestrianobservations.wordpress.com/2011/05/16/us-r...
It seems that $1B/__km__ (not mile, you're right) is sadly par for the course in the US, and Crossrail passes that threshold too. Projects elsewhere are below that threshold though.
2 replies →
If Elon can develop an massively more efficient boring system that the rest of the USA can use; fine, he can have his personal car-way network under his house while I ride around on the highly gridded subway network in Seattle or Austin or Boise. I won't grudge him that.
Also, fwiw, I think cut and cover is a much better approach, despite the street level disruption. Ce la vie.
The question is if he can drop those costs substantially. I certainly wish him the absolute best in doing so. But, if he's going to propose building transportation systems with huge operational costs (SOV tunnels), they had better be efficient in the cost per rider mile and competitive with other technologies.
Except the monoculture here doesn't account for just how terrible buses and trains are.
Mass transit is fundamentally superior and realistically inferior. The second any rational actor has the economic advantage necessary to not ride in a box packed with people of questionable hygiene they'll chose option #2 no matter the negative externalities.
I love this place but people aren't packets in your stream, they actively decide against your well designed "optimal" systems and opt for less efficient "luxurious" options as soon as they possibly can.
Have you ever ridden on a really good train system?
I get the train to work each day and 99% of the time I get to sit down with space to comfortably work on my laptop and I get through a bunch of stuff each way. When I think of the alternative - spending 20% less time but dealing with the risk and frustration of driving and getting nothing done in return - driving myself seems positively toxic.
> The second any rational actor has the economic advantage necessary to not ride in a box packed with people of questionable hygiene they'll chose option #2 no matter the negative externalities.
:: raises hand :: I can easily afford to drive and park. I bus to work instead. No time difference. A number of people are like me.
(the fact that buses allow people who aren't very well house trained on board is a matter that needs to be addressed, no question about it)
I mean, where I lived I _tried_ to use public transit. But with a 50/50 chance of getting passed by the single bus running that route, it just wasn't making sense anymore. And that route ran every 60 minutes .. and also didn't run on Sunday's.
It's not even hygiene. When you ride a train, you cede control of your life to it: you have to plan things by schedule, if it's delayed you are delayed, and you don't even have the luxury of silence. people don't usually vote to let other people have immediate control of their lives like that.
There seems to be this weird thing where people are embracing things that strip autonomy from them.
Reports I read on this a month or more ago suggested that yes, part of his goal is to improve the speed/cost of building tunnels, so that it's feasible to go build a whole bunch of them quickly.
(That being said, building a bunch of tunnels in the bosom of San Andreas strikes me as hubris... But having lived in LA and its traffic for many years, I'm willing to be convinced. :-) )
Hilariously so even. This is basically a subway system, except you have to provide your own seats. All this is succeeding in doing is making Musk look hilariously out of touch.
He may well be out of touch, but at least Musk's suggesting a real solution to current transit problems, even if it's impractical. On the other hand, whenever I read stuff by politicians and engineers with real knowledge and influence in this field, I'm astonished by the total lack of vision.
Even in London, where you have a major, successful transport engineering project (Crossrail) completing, and an entire city dependency on, and supportive of, public transport, you only hear the most anemic plans for future expansion.
For example, TFL is pushing a "New Tube For London" plan that consists of little more than trivial improvements. Some new trains, platform edge doors, etc., scheduled to be delivered in 2050 or so. But absolutely nothing with real ambition, like entirely new lines, that would actually solve London's insane congestion problems.
The problem is not that they cannot imagine the concept of adding new lines, it's just that the costs, both monetarily and politically for doing so, are enormous in the Western world. The eminent domain, permitting, and labor issues around these things paralyzes the development of cities, especially in the United States.
If you want to see ambition in infrastructure, check out China. Being a technocracy bent on having world-class infrastructure, combined with the ability to remove people from their homes for the greater good with little repercussion, they've built some of the most thorough infrastructural improvements ever seen on the planet. For instance, the first line of the Shanghai Metro was opened in 1993; today, it is the longest metro in the world, with 14 lines. And this rapid growth can be seen not just in the megacities like Shanghai, but in countless smaller ones throughout the country—not to mention their incredible construction of their thorough long distance high speed rail network. And on top of that, since their system is so modern, the trains are a hell of a lot safer and more reliable than any piece of infrastructure in the USA.
I think they've made mistakes with how car-oriented the streetscapes are, the generic architectural styles throughout that country, overbuilding before demand arises, and a lack of mixed-use neighborhood zoning, amongst other things; but when it comes to imagining an integrated, efficient commuter rail network, they've killed it.
4 replies →
What? TfL has no ambitions? The crossrail is the biggest construction project in the whole Europe, and it is still not finished and they are already planning for the crossrail 2 that will be started in the next 3-4 years. These are huge projects, much more important than Musk proposal that in London would have been simply ridiculous given that the crossrail alone will move 200 million people per year. If they went for the proposal discussed here instead London congestion would have exploded given the abismally low capacity of this boring company project.
Opening new lines, like building new roads, might hurt everything due to induced demand.
You can reduce demand by not needing everyone to go the same place every day. Why does everyone have to work in London and why don't they already live where they work?
2 replies →
> He may well be out of touch, but at least Musk's suggesting a real solution to current transit problems, even if it's impractical.
He's proposing the transportation equivalent of a personal computer with perfect security, completely crash free, 1000x faster than current PCs that cost lesss tha. We have today. It's not impractical, it's laughable.
> On the other hand, whenever I read stuff by politicians and engineers with real knowledge and influence in this field, I'm astonished by the total lack of vision.
If you want to see total vision, come to a Transportation Research Board annual meeting. Or an APTA annual meeting. Visions are out there. The political support is not.
1 reply →
paralysis via NIMBY is the key problem in the US. many urbanists would be happy to grid every city with subways. we talk about it, wistfully.
but we don't have the political leverage today, and the little we have is with the Democrats, not the Republicans.
"real solution" =! "it's impractical"
Decide which one is it.
Getting to provide your own seats might just be what convinces drivers to use a subway system.
Isn't Musk's track record simply to shoot for the moon (or heck, Mars), then temper things down?
He misses all his deadlines for precisely this reason.
Right now, it's just a video mocked together over two days. It's just a very, very broad "idea".
Things will change over time. Why knock something so early in the development phase?
hilariously.
>But there's no engineering reason why a hole in the ground can't happen.
Well, unless you count damage to whatever happens to be in the way of the tunnel and buildings on top of tunnel. This might be more of a problem in Europe than in the U.S. though.
Look at the Seattle viaduct replacement problem. Bertha came to a standstill for years due to engineers not taking care of a well pipe they thought they removed.
I think the video was just to look flashy and show off the Model 3 design. I imagine most of the tunnel boring will be for mass transit.
> tunnelling is a broadly solved problem. It's difficult, expensive, slow
And now Musk is trying to solve for those three factors
>Musk would do significantly better to drop a full subway network.
In the same way a road offers private and public transportation, a tunnel can do both. Why can't the car be a bus, say?
Because that’s hugely inefficient compared to a subway network.
I live in Paris, France. 100% of its inhabitants have at least one of the 300+ subway stations under 1 km (0.6 miles) of their home. The subway system transports 5M people every single day. That’s twice the city population. That’s also 1.9B per year. During the peak hours you can have up to one train every 90 seconds for a capacity of 700 people each.
> Because that’s hugely inefficient compared to a subway network. I live in Paris, France. 100% of its inhabitants have at least one of the 300+ subway stations under 1 km (0.6 miles) of their home.
That's actually pretty awesome, but you may have noticed that Americans don't really care about efficiency.
Once the tunnel is built, you can put anything you want into it. But I suspect we would still choose cars over trains.
1 reply →
Paris isn't on top of the San Andreas faultline!
1 reply →
Because cars, which are relatively wasteful spacewise, create negative externalities on the public transportation by causing congestion, which resultantly makes public transportation less efficient.
Not an issue in automated underground systems really.
1 reply →
> But there's no engineering reason why a hole in the ground can't happen.
Depending on the specific location, there may very well be. Underground tunnels are vulnerable to earthquake-inflicted damage, for instance, and you would want to avoid repair as much as possible.
Similarly, there may be environmental or conservatory reasons not to build tunnels, or reasons related to cost, or obstacles not yet encountered (that last one is less likely though).
This is Elon's response to a Twitter users similar comment about earthquakes. I don't know enough about physics to say if he's just throwing things out there or if he already did some research.
https://twitter.com/elonmusk/status/824300186834898945
Mmm, so I spoke too soon on that front. In my defense, I've said this a lot to people who are advocates of Hyperloop construction (vacuum tubes are far too easily ruptured to be used in conjunction with transportation), but Musk is correct.
Are subways targets for terrorism the same way that airports are?
2 replies →
Just look at the Big Dig. One small mistake (using the wrong glue or fasteners) can put the safety of the entire safety into question.
Yes, see Pinheiros subway station accident as a good example:
http://oglobo.globo.com/fotos/2007/07/11/11_MHG_sp_metro4.jp... https://www.ita-aites.org/fr/news/download/383_4725e686e12d3...
But those exist for above ground systems too, right? Earthquakes damage roads, bridges. Road and rail impact the environment as well.
The Seattle viaduct is a good example of transportation infrastructure that is very vulnerable to quakes. They're replacing it with a tunnel, so I assume they have worked out how much damage an earthquake will do. At least the tunnel can't fall on anyone, unlike the viaduct which would pancake and render the waterfront inaccessible.
broadly, a generic hole is doable, it's been done many places.
just like software is doable. :)
sometimes you can't do it. but generally... you can. :)
Like Hyperloop?
Hyperloop was a completely infeasible design that Musk quite openly had no intention of building. It mainly seemed to exist as a tactic to attack actual planned public transport expansion in the Bay Area, which Musk compared it unfavourably to based on figures that didn't add up.
I bet this tunnel boring idea was inspired by the lack of an economical path for hyperloop to get to downtown LA. The hyperloop concept took a lot of heat for that.
This is to subways like hyperloop is to high-speed rail. A stupid overelaboration of an existing, proven idea.
Hyperloop is a 'stupid overelaboration of an existing, proven idea'? I was not aware we had so many proven 600mph high speed rails.
7 replies →
Cars and mass transit both have major problems. Cars aren't scalable, and mass transit is often slow when you have to wait for trains/buses to come. (It's worse when you have to change buses or trains.)
The video looks more like a high-speed form of personal rapid transit [1], which is a sort of hybrid system that tries to achieve the best of both. It's like a bus or train system, except the vehicles are smaller and don't operate on a timetable. Instead, they pick you up when and where you want and they take you where you want to go.
[1] https://en.wikipedia.org/wiki/Personal_rapid_transit
edit: tunnelling is a broadly solved problem. It's difficult, expensive, slow, etc
Well, which is it? Solved, or difficult/expensive/slow?
It's all four, They aren't mutually exclusive.
> If he's looking for mega-good, Musk would do significantly better to drop a full subway network.
Agree, except that while many ridiculously wealthy people would be willing to help pay for private access to some tunnels — _perhaps_ subsidizing a system that is later available to others — many fewer wealthy people would help pay for a subway.
What? You could put trains on those tracks.
I think it's neat. I like London, but it'd be sweet to have my car while I was there.
Yeah, like you can't use the car transport floats without a car? Why?
> If he's looking for mega-good
I doubt it. I think he's looking for the same thing as everybody else: mega profits.
Yes Musk only works for the $$$
why not both?
As a counter-point: do some research into Seattle's absolute disaster with tunnelling the last few years. Yes, there are some details of the situation that are specific to Seattle, but it just goes to show that even nowadays, tunnelling isn't as easy as we think it is.